Neural network modeling for small DMFC bipolar plate stack system

Hwa Chaing Chen, Chao Hsing Hsu, Pao Hua Chou, Shu Han Yang, Chi Yuan Chang, Menq-Jiun Wu

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Since the multitudinous parameters make the DMFC model become a complex "black-box" direct methanol fuel cell (DMFC) modeling has become an important and difficult issue of practical experiment. This paper presents the model of the small DMFC Bipolar Plate Stack System. This system is different from the conventional complex mathematical models by the neural network (NN) which is considered the real paramneters of the measurement inputs and outputs. First the training data is the power density of short-term stability tests for the DMFC 10-cell stacks. Then the DMFC model can be obtained by NN method and contented to the different unit cell stacks. Finally, the simulation results in agreement with experimnental results show that the NN modeling method effectively projects the power density on small DMFC packs, such as the developmnent and simulation tool. Therefore the NN modeling method can save much time to reform the conventional mathematical models by the very expensive experimnent.

Original languageEnglish
Pages (from-to)4765-4774
Number of pages10
JournalInternational Journal of Innovative Computing, Information and Control
Volume5
Issue number12
Publication statusPublished - 2009 Dec 1

Fingerprint

Direct methanol fuel cells (DMFC)
Fuel Cell
Network Modeling
Neural Networks
Neural networks
Modeling Method
Mathematical Model
Mathematical models
Cell
Simulation Tool
Black Box
Model
Unit
Output
Modeling
Experiment
Simulation
Experiments

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Information Systems
  • Software
  • Theoretical Computer Science

Cite this

Chen, Hwa Chaing ; Hsu, Chao Hsing ; Chou, Pao Hua ; Yang, Shu Han ; Chang, Chi Yuan ; Wu, Menq-Jiun. / Neural network modeling for small DMFC bipolar plate stack system. In: International Journal of Innovative Computing, Information and Control. 2009 ; Vol. 5, No. 12. pp. 4765-4774.
@article{a107ea43f11e48348e8a8aa0644fac19,
title = "Neural network modeling for small DMFC bipolar plate stack system",
abstract = "Since the multitudinous parameters make the DMFC model become a complex {"}black-box{"} direct methanol fuel cell (DMFC) modeling has become an important and difficult issue of practical experiment. This paper presents the model of the small DMFC Bipolar Plate Stack System. This system is different from the conventional complex mathematical models by the neural network (NN) which is considered the real paramneters of the measurement inputs and outputs. First the training data is the power density of short-term stability tests for the DMFC 10-cell stacks. Then the DMFC model can be obtained by NN method and contented to the different unit cell stacks. Finally, the simulation results in agreement with experimnental results show that the NN modeling method effectively projects the power density on small DMFC packs, such as the developmnent and simulation tool. Therefore the NN modeling method can save much time to reform the conventional mathematical models by the very expensive experimnent.",
author = "Chen, {Hwa Chaing} and Hsu, {Chao Hsing} and Chou, {Pao Hua} and Yang, {Shu Han} and Chang, {Chi Yuan} and Menq-Jiun Wu",
year = "2009",
month = "12",
day = "1",
language = "English",
volume = "5",
pages = "4765--4774",
journal = "International Journal of Innovative Computing, Information and Control",
issn = "1349-4198",
publisher = "IJICIC Editorial Office",
number = "12",

}

Neural network modeling for small DMFC bipolar plate stack system. / Chen, Hwa Chaing; Hsu, Chao Hsing; Chou, Pao Hua; Yang, Shu Han; Chang, Chi Yuan; Wu, Menq-Jiun.

In: International Journal of Innovative Computing, Information and Control, Vol. 5, No. 12, 01.12.2009, p. 4765-4774.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Neural network modeling for small DMFC bipolar plate stack system

AU - Chen, Hwa Chaing

AU - Hsu, Chao Hsing

AU - Chou, Pao Hua

AU - Yang, Shu Han

AU - Chang, Chi Yuan

AU - Wu, Menq-Jiun

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Since the multitudinous parameters make the DMFC model become a complex "black-box" direct methanol fuel cell (DMFC) modeling has become an important and difficult issue of practical experiment. This paper presents the model of the small DMFC Bipolar Plate Stack System. This system is different from the conventional complex mathematical models by the neural network (NN) which is considered the real paramneters of the measurement inputs and outputs. First the training data is the power density of short-term stability tests for the DMFC 10-cell stacks. Then the DMFC model can be obtained by NN method and contented to the different unit cell stacks. Finally, the simulation results in agreement with experimnental results show that the NN modeling method effectively projects the power density on small DMFC packs, such as the developmnent and simulation tool. Therefore the NN modeling method can save much time to reform the conventional mathematical models by the very expensive experimnent.

AB - Since the multitudinous parameters make the DMFC model become a complex "black-box" direct methanol fuel cell (DMFC) modeling has become an important and difficult issue of practical experiment. This paper presents the model of the small DMFC Bipolar Plate Stack System. This system is different from the conventional complex mathematical models by the neural network (NN) which is considered the real paramneters of the measurement inputs and outputs. First the training data is the power density of short-term stability tests for the DMFC 10-cell stacks. Then the DMFC model can be obtained by NN method and contented to the different unit cell stacks. Finally, the simulation results in agreement with experimnental results show that the NN modeling method effectively projects the power density on small DMFC packs, such as the developmnent and simulation tool. Therefore the NN modeling method can save much time to reform the conventional mathematical models by the very expensive experimnent.

UR - http://www.scopus.com/inward/record.url?scp=74349093770&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74349093770&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:74349093770

VL - 5

SP - 4765

EP - 4774

JO - International Journal of Innovative Computing, Information and Control

JF - International Journal of Innovative Computing, Information and Control

SN - 1349-4198

IS - 12

ER -